Skip to content

Portfolio rebalancing

The process of comparing actual weights or risks with targets and ranges, then deciding whether and how to trade after accounting for costs, taxes, liquidity and governance.

Who this is for — Readers with target weights or risk budgets who must decide when a deviation justifies a trade, without confusing risk discipline with a promise of additional return.

Portfolio rebalancing compares actual exposures with objectives, identifies the cause of a deviation and decides whether to move the portfolio toward its targets or ranges. The controlled quantity may be capital weight, risk contribution, duration, delta, currency or another mandate-specific limit.

Rebalancing does not always mean returning to the centre of a range. The decision must weigh control benefits against spread, commission, impact, tax, lot size, liquidity, external cash flows and the risk of trading on noisy estimates. Choosing not to rebalance is also a decision: it lets exposures change with the market.

Rebalancing: target, drift and decision Calendar, tolerance bands and cost-aware optimisation are different policies Rebalancing: target, drift and decision Calendar, tolerance bands and cost-aware optimisation are different policies required trade = target weight − current weight, subject to costs and constraints lower bandtargetupper bandcurrent weight Monitor Prices and cash flows moveweights away from policy. Decide Frequency or band width is agovernance choice, not auniversal threshold. Execute Spread, impact, taxes and lotscan reduce or defer the trade. Verify Post-trade weights, risk andcosts are reconciled. Cyclepedia · educational diagram: state conventions, period and data
Bands separate monitoring from orders: crossing a threshold triggers a documented decision, not always a mechanical return to target.

Why weights change

Before trading, identify the cause:

  • different returns across components;
  • dividends, coupons, contributions or withdrawals;
  • changes in volatility, correlation or sensitivity;
  • maturities, exercises, rolls or corporate actions;
  • an intentional policy change;
  • a pricing, position or reconciliation error.

A breach caused by bad data is not fixed with an order. A weight move can be acceptable within a range, while a material risk change may require action even if capital weights look stable.


Three families of rules

Rule Mechanism Operational advantage Limitation
Calendar reviews or trades on fixed dates straightforward governance may miss large interim drift or trade trivial deviations
Bands acts when a measure leaves a range ties action to deviation requires thresholds, data and rules for simultaneous breaches
Cash-flow led directs contributions, withdrawals or income toward correction may reduce sales and cost depends on flow size and timing

The families can be combined: continuous monitoring, a formal monthly review, priority use of flows and trading only outside bands. No frequency or band width is universally correct. The choice depends on volatility, dependence, cost, tax, liquidity, acceptable risk and governance.


Target, range and destination

A complete policy distinguishes the target; tolerance range; trigger that requires a decision; destination weight; priority when several components breach; and exceptions for closed markets, insufficient liquidity or tax constraints.

The destination may be the target, the range boundary or a cost-aware point. Always returning to the centre can increase turnover; stopping just inside the band leaves less room for another move. The rule should be consistent and tested with realistic costs.


Relationship with strategy

Rebalancing maintains a still-valid strategic policy. An intentional tactical deviation creates a temporary target. Unless the tactical thesis is recorded, the system cannot distinguish an active view from failed execution.

Rebalancing does not create a “bonus” by definition. In some paths it sells a component that rose and buys one that fell; the outcome depends on return dynamics, horizon, dependence and cost. NBER research also documents that predictable flows from large investors can create aggregate market impact in the studied sample. That evidence is not a market-timing rule for an individual portfolio.


Example with a band

Suppose a component has a 50% target and a 45–55% range. After a rally it reaches 56%. With a total portfolio value of €100,000, returning to 50% would require a reduction of about €6,000 before costs and price movements during execution.

The policy might return to 50%, return only to 55%, direct a new contribution to other components, defer the trade when estimated cost exceeds the control benefit, or block execution when data or the market are unreliable. The 5% band is part of the example, not a recommendation.


Costs, taxes and liquidity

Transition cost extends beyond commission. It includes half-spread and slippage, temporary and permanent impact, decision-price shortfall, realised tax, FX, funding and borrow costs, operational time, partial-fill risk and opportunity cost during a slow transition.

CFA Institute notes that thresholds may be wider in taxable portfolios because larger movements may be needed before the risk-control benefit justifies the tax. This is a process consideration, not tax advice; rules and effects depend on jurisdiction and investor.


Control record

Each event should preserve the position-and-price snapshot; target, range and breached measure; diagnosed cause; alternatives; estimated cost and risk; decision owner and timestamp; orders and fills; remaining deviation; and later attribution of costs and effects. This history allows the rule to be evaluated without selecting only favourable episodes.

Common error — Rebalancing more often because monitoring looks more precise. Frequency, turnover and estimate quality form a trade-off; more orders can increase operational risk and cost.


Sources